The hyperpolarization-activated/cyclic nucleotide (HCN)–gated channels make important contributions to neural excitability. In prefrontal cortex, HCN channels are localized on the distal dendrites of layer V pyramidal neurons and decrease neural excitability when they are open. In the present study, using whole-cell voltage clamp recordings, the effect of an arousal peptide, orexin A, on HCN currents in layer V pyramidal neurons from mouse prelimbic cortex (PL), the homolog of the prefrontal cortex was investigated. The results demonstrated that orexin A suppressed HCN currents and shifted their activation curve to a more negative direction. This action of orexin A was blocked by SB334867, an orexin receptor 1 (OXR1) blocker and bisindolylmaleimide, a protein kinase C (PKC) inhibitor, indicating the involvement of OXR1 and PKC. The excitatory effect of orexin A on PL pyramidal neurons was enhanced when HCN currents were diminished, while attenuated when HCN currents were enlarged. In summary, orexin A inhibits HCN currents and enhances excitability of pyramidal neurons in PL, which may contribute to arousal and cognition.
The hypothalamic peptides orexin A and B (also know as hypocretin 1 and 2; de Lecea et al. 1998; Sakurai et al. 1998) play a key role in the persistence of wakefulness (Hungs and Mignot 2001; Sutcliffe and de Lecea 2002; Sakurai 2007), and this role is largely attributed to cortical activation induced by the stimulation of multiple subcortical arousal systems, such as the locus coeruleus, mesopontine and basal forebrain neurons, and the nonspecific thalamocortical projection system (Steriade 2000; Jones 2003). Interestingly, morphological studies have shown that orexinergic projections and orexin receptors (ORs) are also distributed in the cerebral cortex, including somatosensory cortex, motor cortex, and the prefrontal cortex (Lu et al. 2000; Marcus et al. 2001). Additionally, it is reported that orexin A can excite layer 6b neurons of primary somatosensory and motor cortex (Bayer et al. 2004). Our previous studies have indicated that orexin A also shows a direct postsynaptic excitatory effect on pyramidal neurons from the rat prefrontal cortex, and this excitatory process may be involved in the inhibition of whole-cell voltage-dependent potassium currents (Xia et al. 2005). In fact, various ionic pathways underlie the excitatory effects of orexins, such as 1) the activation of sodium–calcium exchanger in γ-aminobutyric acidergic neurons of the arcuate nucleus (Burdakov et al. 2003), 2) the activation of a nonstore-operated calcium permeable channel in the CHO-hOX1-C1 cells (Larsson et al. 2005), 3) the activation of a nonselective cation conductance and blockage of a potassium conductance in nucleus tractus solitarius neurons (Yang and Ferguson 2003) and locus coeruleus neurons (Murai and Akaike 2005), and 4) up-modulation of N-methyl-D-asparate receptor channels (Chen et al. 2008). Together, the responses to orexins may involve several different ionic mechanisms in diverse cellular populations.
The hyperpolarization-activated/cyclic nucleotide (HCN)–gated channel is a mixed-cation conductance with slow kinetics, it is encoded by 4 genes (HCN1–4) and widely distributed in peripheral and central neurons. The HCN channel can be typically activated by membrane hyperpolarization and induce an inward cation currents—HCN currents (also know as Ih currents). The reversal potential of HCN currents is from −25 to −40 mV (Robinson and Siegelbaum 2003). HCN channels contribute to a wide range of physiological functions, including cardiac and neuronal pacemaker activity, the setting of resting potentials (RPs). Recently, many studies have demonstrated that the HCN channels are also critical for the regulation of neuronal excitability in hippocampus and prefrontal cortex (Day et al. 2005; Arnsten 2007; Carr et al. 2007; Wang et al. 2007). In these areas, HCN channels are mainly located on the dendritic spines of pyramidal neurons (Magee 1998; Santoro et al. 2000; Lorincz et al. 2002; Notomi and Shigemoto 2004; Yu et al. 2004; Nusser 2009). Given that dendritic spines are important in regulating the neuronal excitability and synaptic plasticity, it is believed that HCN channels are involved in the integration of excitatory synaptic input in these neurons and thereby influences the excitability of neural network (Magee 1999; Williams and Stuart 2003; Day et al. 2005; Oviedo and Reyes 2005; Shin and Chetkovich 2007; George et al. 2009). Importantly, some transmitters and drugs can influence the neuronal excitability via their regulation on HCN currents. For example, 1) brief application of dopamine causes a reduction in the excitability of layer V entorhinal cortex pyramidal neurons, which can best be explained by an increase in HCN channels conductance (Rosenkranz and Johnston 2006); 2) stimulation of α2 receptors by noradrenaline (NA) increases the activity of prefrontal cortex neurons during delay periods of working memory, demonstrates by voltage-clamp analysis that α2-NA receptor stimulation inhibits HCN currents (Carr et al. 2007; Wang et al. 2007); 3) some anticonvulsant drugs, like lamotrigine, can selectively reduce the excitability of hippocampal CA1 pyramidal neurons by upregulation of HCN currents (Poolos et al. 2002); and 4) the sleep-modulating peptide cortistatin can augment the HCN currents in hippocampal neurons (Schweitzer et al. 2003). Orexin A, as a classical wake-promoting neuropeptide, can enhance the activity of pyramidal neurons in prefrontal cortex effectively, but whether orexin A could regulate HCN currents of pyramidal neurons in prefrontal cortex is still little known.
In the present study, using whole-cell patch clamp recordings on layer V pyramidal neurons of mouse prelimbic cortex (PL) that is one important part in the mouse homolog of the prefrontal cortex, the regulation of HCN currents on neuronal excitability is first observed, and then the interaction between orexin A and HCN currents is systematically studied. The results indicated an inhibition effect of orexin A on the HCN currents, which may enhance excitability of pyramidal neurons in prefrontal cortex, and consequently promote arousal and cognitive function.
Materials and Methods
All experiments were carried out in accordance with China Animal Welfare Legislation and were approved by the Third Military Medical University Committee on Ethics in the Care and Use of Laboratory Animals. Kunming mouse (P14–21) was rapidly decapitated, and then the brain was quickly removed and submerged in a 0 °C sucrose solution containing (mM): sucrose, 220; KCl, 2.5; Na2HPO4, 1.25; NaHCO3, 26; MgCl2, 6; CaCl2, 1; glucose, 10. Coronal slices (300–350 μm), including the ventral anterior cingulated, infralimbic, and PL cortices (Paxinos and Watson 1998; Fig. 1A), were prepared using an oscillating tissue slicer (Leica, Wetzlar, Germany, VT1000) and transferred to a holding chamber containing artificial cerebrospinal fluid (aCSF) for a minimum of 90 min at room temperature (22–24 °C) prior to recording. aCSF containing (mM): NaCl 124; KCl 3; NaHCO3, 26; MgCl2, 2; CaCl2, 2; glucose 10. Slices were then transferred to a submersion-type recording chamber and perfused at a rate of 1–2 mL min−1 with aCSF. All aCSF solutions were constantly aerated with a mixture of 95% O2–5% CO2 to maintain pH ∼7.4.
Whole-Cell Clamp Recordings
Whole-cell current clamp recordings in brain slice neurons were made as described previously (Chen et al. 2008). Layer V pyramidal neurons in PL were targeted for recording using an upright microscope equipped with Leica infrared-differential interference contrast (IR-DIC) optics, a ×40 water immersion objective, and an video imaging camera. Patch pipette (3–7 MΩ) filled with internal solution containing (mM): potassium gluconate 125; KCl, 20; 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, 10; ethyleneglycol-bis(2-aminoethylether)-N,N,N',N'-tetra acetic acid, 1; MgCl2, 2; ATP, 4; pH 7.2–7.4. The pipette resistance, as measured in the bath, was typically 4 ± 0.5 MΩ. After gigaohm seal formation and patch rupture, neurons were given at least 5 min to stabilize before data were collected. Following patch rupture, series resistance was compensated 50–70% and continually monitored throughout the experiment. Cells were discarded if series resistance increased by >15%. The signal was amplified using an EPC10 amplifier (HEKA Elektronik, Lambrecht/Pfalz, Germany) and stored for off-line analysis with Pulse/Pulsefit v.8.74 (HEKA Elektronik) and Igor Pro v.4.03 (WaveMatrics). Recordings were performed at room temperature unless stated otherwise. Additionally, some cells were intracellularly labeled with biocytin (0.5%) to confirm identification.
Animals were anesthetized deeply with 3% pentobarbital (1 mL/kg, intraperitoneal) and then were perfused through the ascending aorta with 10 mL of 0.9% saline, followed by 50–60 mL of phosphate-buffered 4% paraformaldehyde in 0.1 M phosphate-buffered saline (PBS), pH 7.0. Brains were removed, postfixed for 4 h in paraformaldehyde, and then allowed to equilibrate in 30% sucrose in 0.1 M PBS (Sigma, St Louis, MO). Brains were then sectioned coronally on a freezing microtome and stored in PBS at 4 °C. Prefrontal sections (cut at 10 μm by a freezing microtome) were selected from each brain and stained by immunofluorescence for ORs using routine methods. Briefly, for immunostaining, the coronal sections were first incubated with goat anti-ORs antiserum (1:100) with 3% normal bovine serum (Santa Cruz, CA) for 2 days at 4 °C and then washing with PBS. A fluorescein isothiocyanate labeled antibovine secondary antiserum (Santa Cruz, CA) was used at a dilution of 1:500. ORs-immunoreactive cells were visualized by fluorescence microscope (Olympus, Tokyo, Japan).
All reagents were obtained from (Sigma) with the exception of ZD7288 and SB334867 (Tocris Bioscience, Bristol, UK). Orexin A was added by puff application to neurons. Other drugs were applied by switching perfusion from aCSF to a solution containing the desired drug. The methods for puff application have been previously described (Sutcliffe and de Lecea 2002). In brief, Orexin A was ejected by pressure from a micropipette (tip diameter: 2–5 μm; duration of application 10–15 s) positioned adjacent to the recording microelectrode. Puff application of aCSF (without Orexin A) with a similar duration did not influence the membrane potential of the recorded neurons. Ba2+ was bathed in at a concentration (200 μM) that blocks Kir2 channels in order to leave HCN currents intact. All drugs were prepared as concentrated stock solutions and frozen at −20 °C until use.
All quantitative values are given as the mean ± standard error of the mean (SEM). Statistical comparisons were made by using the unpaired or paired Student's t-test with SPSS (version 13) software, as appropriate. A P value of <0.05 was considered statistically significant.
Whole-cell recordings were obtained from a total of 122 layer V pyramidal cells located within the mouse prefrontal cortex PL area. These neurons were identified using IR-DIC video microscopy by their large (≥20 μm diameter) pyramidal-shaped cell body with long apical dendrite extending toward the pial surface (Fig. 1B). A subset of these cells was injected with biocytin for further morphological identification. All the labeled cells had morphological features of pyramidal neurons as described previously: a pyramidal soma and a prominent apical dendrite (Fig. 1C). The mean resting membrane potential of neurons recorded in current clamp was –65 ± 4.6 mV. All of these neurons had action potentials when evoked by 200 ms depolarizing currents (Fig. 1D), and most of them showed spike frequency adaptation in response to depolarizing current injection (Fig. 1E).
HCN Channels Regulates the Excitability of Pyramidal Neurons in PL
A series of reports have indicated that HCN channels are localized on the distal dendrites of layer V pyramidal neurons in neocortex and appear to shunt synaptic inputs when they are open (Robinson and Siegelbaum 2003; Wahl-Schott and Biel 2009). In the present study, layer V pyramidal neurons of PL area were first examined for electrophysiological features of HCN currents. Injection of negative current steps (−350 to 0 pA in 50-pA steps) produced a rapid hyperpolarization, followed by a depolarizing sag (Fig. 2A, n = 36). This response pattern is typical of neurons expressing HCN channel currents. The amplitude of the voltage sag increased with negative somatic current injections. Meanwhile, the whole-cell voltage clamp recordings showed that HCN-like inward currents were activated by a series of hyperpolarized voltages (−60 to −120 mV, 10 mV step), which are typical for HCN channels (Yu et al 2004). The amplitude of the HCN-like current at −120 mV was 115.7 ± 10.5 pA (Fig. 2B, n = 36), with the leak currents controlled at ±3.0 pA. The small somatic HCN-like currents recorded in these neurons may reflect current generation at sites distant from the soma (as in dendritic spines) and subsequent signal attenuation resulting from a large dendritic electrotonic length (Destexhe et al. 1996). To further examine the existence of HCN currents in PL pyramidal neurons, the protocol was repeated in the presence of the HCN channels blocker ZD7288 and HCN currents agonist 8-Br-cAMP. ZD7288 (50 μM) or 8-Br-cAMP (1 mM) was, respectively, applied to the bath for 10 min. The HCN-like steady-state current amplitudes were normalized to the amplitude of HCN-like currents in control conditions. At voltage clamp recordings of −120 mV, applying ZD7288 significantly decreased the amplitude of HCN-like currents (decreased by 73.7 ± 6.8%, P < 0.01, n = 7; Fig. 2C,D). Conversely, the amplitude of HCN-like currents was elevated during application of 8-Br-cAMP (increased by 43.4 ± 5.3%, P < 0.05, n = 5; Fig. 2C,D). Taken together, we conclude these HCN-like currents are HCN currents.
In prefrontal cortex, HCN channels localized on the distal dendrites can influence somatic excitability and spike firing (Carr et al. 2007; Wang et al. 2007; Winograd et al. 2008). Here, this phenomenon was examined by blocking HCN channels in mouse layer 5 pyramidal neurons of PL area. The membrane potential was initially adjusted to −50 ∼ −53 mV using intracellular injection of direct current, and the recording neurons (n = 6) fired spontaneously under this condition. Bath application of ZD7288 produced a modest depolarization in the membrane potential and dramatically increased spike firing (Fig. 3A). The increased firing activity did not completely recover to the control (predrug) level even after 10 min washout. In another group (n = 6), the excitability was measured as the number of action potentials evoked by a fixed amplitude of current injection (1 s duration) from RP (−62 ∼ −68 mV), the current amplitude (+50 pA) was chosen that evoked 3 to 5 action potentials. During the period of drug application, ZD7288 induced a significant increase in the average number of spikes per depolarizing pulse (Fig. 3B).
Subsequently, changes of neuronal excitability were observed when HCN currents were enhanced by 8-Br-cAMP in PL pyramidal neurons. The membrane potential was initially adjusted to −46 ∼ −50 mV so that PL pyramidal neurons (n = 5) fired at much higher frequency. Bath application of 1 mM 8-Br-cAMP dramatically decreased the spike firing (Fig. 3C). The decreased firing activity could recover to control levels after 10-min washout. Accordingly, in another group (n = 5), excitability was measured as the number of action potentials evoked by a fixed amplitude of current injection (+70 pA), the data indicated that 8-Br-cAMP induced a significant decrease in the number of spikes per depolarizing pulse (Fig. 3D).
Orexin A Excites PL Pyramidal Neurons
Previous study reported that orexin A exerts postsynaptic excitatory effects on the acutely isolated prefrontal cortex pyramidal neurons, and this excitatory effect displayed dose dependence (1–10 μM; Xia et al. 2005). In the present study, excitatory effect of orexin A on PL pyramidal neurons was studied in slices. Whole-cell current clamp recordings from PL pyramidal neurons show that superfusion of orexin A (400 nM, n = 42; 200 nM, n = 18; Fig. 4C) produced a rapid, sustained depolarization, accompanied by a rapid firing of action potentials. After washout of orexin A, the membrane potential and action potential frequency returned to control levels. This data indicated that orexin A had dramatic excitatory effect on PL pyramidal neurons at concentration of 400 nM, and the excitatory effect of orexin A still remained when the concentration dropped to 200 nM. Furthermore, the excitatory action of orexin A on PL pyramidal neurons persisted in the presence of tetrodotoxin (500 nM) that was known to block synaptic transmission, and was attenuated by the selective orexin receptor 1 (OXR1) antagonist SB334867 (1 μM, Fig. 4D). Accordingly, the results of immunofluorescence revealed that OXR1s were mostly located on cell bodies and dendritic process of layer V pyramid neurons in PL area (Fig. 4A). While, the distribution of OXR2s in PL was relative scarce compared with OXR1s and mostly located on cell bodies of layers V pyramid neurons (Fig. 4B).
Orexin A Suppresses HCN Currents
The data above indicated that blockade of HCN channels could enhance the activity of layer V pyramidal neurons in PL. On the other hand, these neurons could also be excited by orexin A. Therefore, exploring the interaction between orexin A and HCN channels became a key point in the present study. Orexin A (400 nM) was superfused before recording of HCN currents, and barium was bathed in aCSF at a concentration of 200 μM that blocked Kir2 channels in order to leave the HCN currents intact. Applying orexin A produced a reduction of voltage sag in recording cells in response to hyperpolarizing current steps (−350 to −150 pA, in 50 pA steps; Fig. 5A), and the voltage sag ratio (peak voltage change/voltage change at steady state) induced by HCN currents was significantly decreased after application of orexin A (from 1.31 ± 0.09 to 1.08 ± 0.07, P < 0.05, n = 7; Fig. 5B). Correspondingly, current traces were evoked by a 1-s voltage ramp from −70 to −120 mV (Fig. 5C,D; control: black traces), and orexin A produced a significant decrease in the amplitude of HCN currents evoked by this voltage-ramp protocol (Fig. 5C,D, light gray traces). The suppression of orexin A on HCN currents was displayed in almost all hyperpolarized voltage steps (Fig. 5E, I–V:−70 to −120 mV, n = 14). After washout of orexin A for 10 min, the amplitude of HCN currents recovered to the control level (Fig. 5C,D; recovery: gray traces), indicating that the inhibitory effect of orexin A on HCN currents was reversible.
Furthermore, we continued to observe the influence of orexin A on HCN currents activation curve in PL pyramidal neurons. Cells were held at −60 mV, and HCN currents were allowed to be partially activated by hyperpolarizing test commands within the range −120 to −60 mV in 10 mV increments. The cells were then subjected to a voltage jump to −120 mV to obtain full activation of HCN currents (Fig. 6A). The tail currents were shown in detail in Figure 6B. The activation curve was calculated by using the Boltzmann equation Itail.max-Itail/Itail.max (Tanabe 2007, each point represents the mean ± SEM of 8 cells; Fig. 6C). The activation curves of HCN currents were generated before and after superfusion of 400 nM orexin A, which revealed a negative shift of the half-activation voltage (V1/2) from −83.8 ± 1.1 to −91.5 ± 2.7 mV.
To further examine the inhibitory effect of orexin A on HCN currents in PL pyramidal neurons, the protocol was repeated in the presence of the OXR1 antagonist SB334867 at −120 mV. As indicated in Figure 7A, the inhibitory effect of orexin A on HCN currents was partially attenuated after bath application of SB334867 for 5 min. The amplitude of HCN currents was decreased to 51.6 ± 6.7pA by 400 nM orexin A and then recovered to 75.4 ± 8.2pA by preapplication of 1 μM SB334867 (P < 0.05, n = 3; Fig. 7B).
Previous study has demonstrated that orexin A excites prefrontal neurons by activation of phospholipase C and protein kinase C (PKC) pathways (Song et al. 2005). Furthermore, the activation of PKC can inhibit HCN currents in some brain areas. For example, 1) neurotensin inhibits HCN currents via activation of PKC in the substantia nigra (Cathala and Paupardin-Tritsch 1997); 2) serotonin reduces HCN currents in dopaminergic ventral tegmental area neurons by acting at serotonin 5-HT2 receptors, which activate PKC (Liu et al. 2003); and 3) in hippocampal CA1 pyramidal neurons, 3 Hz synaptic stimulation downregulates HCN currents via activation of group 1 metabotropic glutamate receptors and subsequent stimulation of PKC (Brager and Johnston 2007). Therefore, PKC activation as being necessary for the downregulation of orexin A on HCN currents was tested. HCN currents steady-state current amplitudes at different time point after incubation of PKC inhibitor BIS-II were normalized to the amplitude of HCN currents in baseline conditions. After bath incubation with BIS-II for from 30 to 90 min, the inhibitory effect of orexin A on HCN currents was gradually attenuated at −120 mV (Fig. 7C). As indicated in Figure 7D, the amplitude of HCN currents was reduced to 52.6 ± 4.8% by 400 nM orexin A. After incubation with 1 μM BIS-II, the amplitude of HCN currents, respectively, recover to 65.4 ± 5.5% at 60 min (P < 0.05, n = 7) and to 83.6 ± 6.7% at 90 min (P < 0.01, n = 5).
The State of HCN Channels Influences Excitatory Effect of Orexin A on PL Pyramidal Neurons
Layer V pyramidal neurons that had no spike firing at RP in PL were selected. These neurons could be excited by 400 nM orexin A (Fig. 8A). 50 μM ZD7288 was bathed in recording solution for 10 min, and then the excitatory effect of orexin A on recording cells was observed again. The results indicated that the spike firing induced by orexin A was enhanced obviously after application of ZD7288 (from 3.84 ± 0.73 Hz to 5.62 ± 0.95 Hz, P < 0.05, n = 8; Fig. 8). The recording cells had slight spike firing during application of ZD7288 alone (0.58 ± 0.26 Hz, n = 8, results not shown). Subsequently, the change of excitatory effect of orexin A on PL pyramidal neurons was observed when HCN currents were augmented by application of 8-Br-cAMP. The neurons that could be excited by orexin A effectively were selected. Then, 1 mM 8-Br-cAMP was bathed in recording solution for at least 10 min before orexin A application again. As indicated in Figure 8, the spike firing induced by orexin A was attenuated significantly compared with the former (from 4.41 ± 1.22 Hz to 2.38 ± 0.83 Hz, P < 0.05, n = 7).
The studies reported here provide evidence that HCN channels have powerful influence on the firing properties of PL pyramidal neurons. Blockade of HCN channels increases the activity of PL pyramidal neurons; conversely, upregulation of HCN currents dramatically decreases the activity of PL pyramidal neurons. Most importantly, the present studies indicate that orexin A can decrease the amplitude of HCN currents and shift their activation curve to a more negative level in PL pyramidal neurons, this effect is mediated by OXR1 and PKC signaling pathway. Furthermore, the excitatory effect of orexin A on PL pyramidal neurons is enhanced when HCN currents are diminished, while attenuated when HCN currents are enlarged. Together, these results imply a functional interaction between orexin A and HCN channels in modulation of prefrontal cortex neuronal activity.
HCN Channels Regulates Neuronal Excitability of Pyramidal Neurons in Prefrontal Cortex
It is believed that HCN channels distributed in layer V pyramidal neurons of prefrontal cortex play an important role in the process of dendritic integration (Robinson and Siegelbaum 2003). Thus, HCN channels can participate in the execution of many complex functions in prefrontal cortex, such as working memory (Wahl-Schott and Biel 2008). As is known, the PL area has a much closer relation with these functions of prefrontal cortex (Heidbreder and Groenewegen 2003; Nasif et al. 2005). But, the details of HCN channels in PL area are still little known. In the present study, HCN channels were rapidly activated by hyperpolarizing voltage steps in deep-layer pyramidal neurons of PL. The property of HCN currents was in agreement with single-cell RT-PCR profiling showing that pyramidal neurons in prefrontal cortex expressed high levels of HCN1 mRNA (Day et al. 2005). Channels dominated by HCN1 subunits are rapidly gating and have relatively depolarized activation voltage dependence (Chen et al. 2001; Wang et al. 2001) like that seen in prefrontal cortex (Franz et al. 2000; Ulens and Tytgat 2001). However, other HCN subunit mRNAs were also detected in pyramidal neurons of prefrontal cortex, particularly HCN2 mRNA (Day et al. 2005).
Previous studies of HCN currents with properties like those seen in prefrontal cortex suggest that they are important determinants of neuronal excitability (Day et al. 2005; Arnsten 2007; Carr et al. 2007; Wang et al. 2007; Winograd et al. 2008). Increased HCN currents decreases intrinsic and synaptic excitability through a current leak path. Conversely, blockade of HCN channels enhances both excitability and excitatory postsynaptic potential responses, which can be attributed to an increase in membrane resistance. Here, the influence of HCN currents on firing properties of PL pyramidal neurons was directly observed. Our results confirmed the important role of HCN channels in regulating the neuronal excitability in PL. Blockade of HCN channels increased the firing rate of PL pyramidal neurons; adversely, the firing rate was decreased. These results were consisted with those described in the process of working memory, blockade of HCN channels strengthens delay-related firing of prefrontal cortex pyramidal neurons (Wang et al. 2007).
Orexin A Suppresses HCN Currents, Which Is Mediated by PKC Pathway
As discussed above, HCN channels regulate neuronal excitability, and many transmitters and drugs can influence the neuronal excitability via their regulation on HCN currents (Poolos et al. 2002; Rosenkranz and Johnston 2006; Carr et al. 2007). Here, the results demonstrated that orexin A suppressed HCN currents of pyramidal neurons in the PL area. This was based on several observations. First, the effect of orexin A in current clamp recordings (decreased voltage sag) was very similar to the effects of HCN channel blockade using Cs+ or ZD7288 (Magee 1998; Berger et al. 2001). Second, the orexin A-sensitive currents evoked by hyperpolarizing steps in voltage clamp exhibited similar kinetic tendency to the currents that are blocked by the HCN channel blocker ZD7288, and the amplitude of HCN current decreased by orexin A was comparable to that decreased by NA in prefrontal cortex (Carr et al. 2007). Third, orexin A shifted the activation curve of HCN current to a more negative level in PL pyramidal neurons. Since the HCN1 isoform exhibits a more depolarized membrane potential for activation than the HCN2 isoform (Robinson and Siegelbaum 2003), preferential inhibition of HCN1 could result in a negative shift of the half-activation potential of HCN current.
In addition, the best well-characterized signaling pathway for regulating HCN channels depends on the intracellular cAMP level (Robinson and Siegelbaum 2003). But, some studies in substantia nigra, CA1, and prefrontal cortex have provided evidence that downregulation of HCN currents also required PKC activation (Cathala and Paupardin-Tritsch 1997; Liu et al. 2003; Brager and Johnston 2007). Blockage of PKC could significantly decrease membrane resistance, increase voltage sag of HCN currents, and therefore decrease neural firing frequency (Brager and Johnston 2007). In our previous study, we demonstrated that the excitatory effect of orexin A on prefrontal pyramidal neurons was mediated by activation of PKC signaling pathways, and blockage of PKC could inhibit excitatory action of orexin A effectively (Song et al. 2005). Here, the present results also indicated that the suppression of orexin A on HCN currents was attenuated by application of PKC inhibitor BIS-II. Additional studies are needed to determine whether PKC acts directly on HCN channels or via an intermediary, such as the transactivation of a protein tyrosine kinase (Shah and Catt 2004). Together, the orexin A–HCN modulation process in prefrontal cortex may be clarified: Stimulation of OXR1s in prefrontal cortex by orexin A, activates PKC, thus inhibits HCN currents. Additionally, the results of immunofluorescence in PL pyramid neurons reveal that OXR1s are also located in the dendritic processes where HCN channels are densely distributed. Although further morphological evidences would be helpful, these results provide a cellular basis for the interaction between orexin A and HCN channels.
The Modulation of Orexin A–HCN Signaling on Neural Excitability of Prefrontal Cortex
It is reported that orexin A can only excite layer 6b neurons of primary somatosensory and motor cortex in rat (Bayer et al. 2004). Our previous studies have indicated that orexin A also shows a direct postsynaptic excitatory effect on acutely dissociated pyramidal neurons from the rat prefrontal cortex (Xia et al. 2005). In the present study, our morphological result further demonstrated that OXR1 were distributed in every layer of mouse PL area, especially located on layer V pyramidal neurons. The electrophysiological data also shown that orexin A could obviously excite layer V pyramidal neurons in a dose-dependent manner, and this excitatory effect of orexin A could be blocked by SB334867, an orexin OXR1 blocker. Therefore, it is believed that orexin A has a different excitatory effect in prefrontal cortex compared with other areas of cerebral cortex.
Here, the present study further demonstrates that orexin A can suppress HCN currents effectively. Given the importance of HCN currents in shaping neural activity, it will provide a rational electrophysiological basis for the excitatory effect of orexin A on pyramidal neurons in the PL area. In other words, the excitatory effect of orexin A on PL pyramidal neurons may involve its inhibition on HCN currents. Accordingly, in our study, the excitatory effect of orexin A on PL pyramidal neurons following the state alteration of HCN channels has also changed, it can be attenuated when HCN currents are augmented by 8-Br-cAMP and enhanced when HCN currents are diminished by ZD7288. In our study, ZD7288 could effectively inhibit HCN currents but not completely block HCN channels (as indicated in Fig. 2C,D) at RP (different from the subthreshold potential in Fig. 3A); thus, orexin A could further inhibit HCN currents in the presence of ZD7288. Nonetheless, the possibility that the effects of orexin A on neural excitability and HCN channels may be independent cannot be excluded in the present study. Although the direct evidences for orexin A–HCN modulation on neural excitability of prefrontal cortex are insufficient, the model of orexin A–HCN modulation is still an ideal pathway for dynamically regulating the strength of prefrontal cortex excitability.
The Role of Orexin A–HCN Modulation on Arousal and Cognitive Function of Prefrontal Cortex
Orexins play a critical role in regulating arousal state, their deficiency can cause narcolepsy in humans and animals (Chemelli et al. 1999; Hara et al. 2001; Peyron et al. 2001). The prefrontal cortex is a brain region whose activity (such as working memory and attention) is correlated with level of wakefulness (Hofle et al. 1997; Paus et al. 1998; Thomas et al. 2000; Lambe and Aghajanian 2003). In narcolepsy, the specifically “executive” aspects of attention of prefrontal cortex are affected (Naumann et al. 2001; Rieger et al. 2003). Furthermore, a recent study has reported that both sporadic narcoleptic dogs and human narcolepsy–cataplexy subjects showed a significant decrease of OXR1 expression in frontal cortex (Mishima et al. 2008). This phenomenon is possibly due to long-term postnatal loss of ligand production in narcoleptic subjects. As is known, OXR1 has a significantly higher affinity for orexin A (Sakurai et al. 1998). Thus, it is believed that orexins, especially orexin A, play important role in prefrontal cortex. But, the details of cellular effects of orexin A are still little known in prefrontal cortex, our previous study have indicated that orexin A shows a direct postsynaptic excitatory effect on pyramidal neurons from the prefrontal cortex. Here, the present study further provides a model of orexin A–HCN modulation on neural excitability in prefrontal cortex.
Additionally, it has been reported that orexin B has a presynaptic excitatory action on the terminals of the nonspecific thalamocortical projection system in the medial prefrontal cortex and can elicit a dramatic spontaneous excitatory postsynaptic current in layer V pyramidal neurons, which enhances cortical activation (Lambe and Aghajanian 2003; Liu and Aghajanian 2008). In the present study, the modulation of orexin A on HCN channels may provide another direct postsynaptic pathway for dynamically regulating the strength of prefrontal cortex excitability during wakefulness. These modulations of orexins on prefrontal cortex neurons including both “presynaptic orexin B-thalamocortical terminals” and “postsynaptic orexin A–HCN channels” may be important for the maintenance of a vigilant working circumstance in prefrontal cortex, based on which higher function (such as “attention” and “working memory”) of prefrontal cortex could be smoothly completed during wakefulness.
National Natural Foundation of China (30800450); Medicine and Health Scientific Foundation of People's Liberation Army (06MB237); the National Basic Research Program of China (No.2006CB500800).
We thank Dr Marion Weiss for grammar comments on this manuscript. Conflict of Interest: None declared.